Seminar Series

FE Real-Time Simulation with Conventional Hardware Tools – Elements, Algorithms, and Examples


Prof. M. W. Zehn (1) & Dr. D. Marinkovic (2)


(1) Department of Structural Analysis, Berlin Institute of Technology, Germany. (2) Faculty of Mechanical Engineering, University of Nis, Serbia

Time & Place

Thu, 08 Mar 2018 14:00:00 NZDT in E7 (Engineering Core Bldg)

All are welcome


An important ingredient of computer aided engineering (CAE) tools is the simulation of physical processes and phenomena. Finite Element Method (FEM) has deservedly established itself as the method of choice for general deformable structures. The conventional approach implies off-line simulations followed by the assessment and analysis of the obtained results in postprocessors. However, the strong development of advanced visualization systems, followed by further development of already existing software components and necessary hardware, have enabled the concept of virtual reality and real-time simulation. Interactive simulation of deformable objects has become increasingly significant in various applications ranging from entertainment industry to virtual reality simulators and to engineering tasks in the field of Multi-Body System (MBS) dynamics or in Biomechanics. Simulation at interactive frame rates implies real-time or nearly real-time computation and graphical representation of modelled deformable objects.
This presentation gives an overview of our developed FEM formalisms with the objective of enabling highly efficient, interactive simulations with geometrical nonlinearities included. The formalisms range from mass-spring systems to modal-space based solutions and co-rotational FEM formulations for solids and shells. Real-time simulation of dynamical behavior of deformable objects with geometrically nonlinear effects included is a challenging task. The authors have developed FEM (finite element method) formalisms in an attempt to balance the requirements for high numerical efficiency (real-time) and numerically rather demanding simulation (a large number of degrees of freedom and nonlinearities involved). The main ideas of the development are presented.